The Skittles Project Final

The Skittles Project
Angela Clayton
The following is a statistical analysis, to uncover the mystery of how many Skittles of each color
are in each bag. Weve taken a sample of 49 bags of skittles, containing 2,997 individual Skittles
candies to unlock the answer and to understand critical problem solving steps through
Statistical mathematics equations.
Skittles Sample Data Collection:

The above is a snapshot of our overall data gathering. This shows the results from the total
number of bags 49.
It was interesting seeing the totals from the class. I think that you can see above from my
numbers that there were a few differences in the overall outcomes. Had my bag held true for
the entire sample, there would have been far less total yellow and purple than total orange,
and it the classs results we see that orange overall, was one of the highest reported colors per
bag. I think this is a great example of why you need to make sure that you arent under
sampling the population.
5 Number Summary for Skittles Sample Data Set:

(a) mean number of candies per bag- 61.2
(b) standard deviation of the number of candies per bag- 8.9
(c) 5-number summary for the number of candies per bag:
Minimum: 52 First Quartile: 58 Median: 60 Third quartile: 62 Maximum: 110
The total candies in each bag ended up being much more interesting that I thought that it
would be. I didnt think that there would be so many outliers. My bags sample fell closer in line
with the class mean. I think overall the graph is showing skewed right, but I think the major
outlier was someone used a bigger bag then the rest of the class.
Categorical data (also known as qualitative data) can be put into a number of different groups
or categories, and may not have any logical or sequenced numbering. Quantitative variables
can be ordered and measured in a logical or sequential way. Both types of data are often
necessary to perform a statistical analysis or test. Categorical data in our skittles examples
would be the color of skittle, while the quantitative data would be the number of skittles of
each individual color in the bag.
Categorical data is best fit into a binned frequency chart, pie charts, and bar graphs. All of these
charts or graphs provide way for the categorical data to be represented so that the quantitative
data associated with each is categorical data set is represented in a meaningful way.
Quantitative data is best fit into stemplots, histograms, and boxplots just to name a few. The
reason these are good fits for qualitative data (without any categorical data associated to them)
is because they help organize and structure the quantitative data into a logical or sequential
order that provides a structure to adequately analyze the data set.
Group members began the project by estimating the proportion of each color of original skittles
within the overall sample gathered by the class. Our consensus was that the proportions in the
class sample should be approximately equal to each individuals sample (i.e. the proportion of
skittle colors in the individuals 2.17 ounce bag). Thus, according to our estimate, each color
would then amount to about 17-20% of the total sample in one bag.
A random sample occurs when a sample size, n, from a population of size, N, has an equal
chance of occurring. In other words, a random sample represents a sample representative of a
population because of the role of chance or randomness plays in obtaining the sample. The
population in this project includes all original skittles in 2.17 ounce bags in the United States.
Despite the fact that the class sample includes 2,997 skittles (or 49 2.17 ounce bags), we need
more individuals to conclude with some certainty that the results can be extended to the entire
population. Therefore, the class sample does not represent a random sample. In addition, it is
unclear where each participant is purchasing his/her skittles. Many students may reside in Salt
Lake City, but others may be in other counties and even other states. This factor also makes it
difficult to expand our findings to the population and may even skew or make it unclear what
the population is.
Confidence Intervals:
A confidence interval is a range of values or numbers that can help you understand (with x% of
confidence) of where the population mean resides. For example, if I have a 99% confidence
interval, I am 99% confident that the population mean resides within the range (ex. Lower
Bound-50, Upper Bound-60). What this means is that the other 1% is expected to not have the
population mean. So that would mean there is 1% confidence that the population mean would
lie outside of the 50,60 range. The higher you want your confidence interval to be, the wider
the range will have to be. If you want your range to be narrower, you will have to have a lower
confidence interval to say 75%.
/2 (1) is the formula we use to find the interval for the population
proportion. Here, is 1% so /2 = .05%. The Z value of .05 = 2.576. p = 566/2997 = .189.
.189 2.576 . 189(1 .189) 2997 = (.171, 2.07)
This means that we can be 99% confident that the population proportion is between .171 and
2.07.
/2 is the formula we use to find the interval for the population mean. We put all of
the results for skittles in a big into a list in the calculator, and did a 1 variable test on the list. S =
8.868, n = 49, x = 61.163. The t value of .025, /2, at 48 degrees of freedom, n-1, is 2.011.
61.16. 2.011 8.868 7 = (58.615, 63.711)
This mean we can be 95% confident that the population mean is between 58.615 and 63.711.
( (1) 2 2 2 , (1) 2 21 2 ) are the two formulas we used to find the interval of
the standard deviation. We used the same s = 8.868 and n = 49. /2 = . 01. The x^2 value of .01
= 73.683. And the x^2 value of .99 is 142.386. These are both with 48 degrees of freedom.
Plugging those values into our interval formulas and taking the square root of them yields the
result: (7.158, 11.933)
This means that we can be 98% confident that the standard deviation estimate is between
7.158 and 11.933
Reflection:
The Skittles Statistical analysis project provided a very unique way to engage into real world
situations. I am an avid basketball fan, and there is always talk about statistics that go incredibly
in depth and analytics associated with those statistics. I think that this project really made me
look at some of the stats that are brought up and that analytics involved. I have a better grasp
of what stats are truly important and can see how some can be twisted to make a point rather
than being true. While this is great for my hobbies, stats are incredibly important in my future
career.
I plan on moving on and majoring in Pharmacy, which will have a large number of statistics that
I am both presented with and that I am also going to need to develop. This class and project has
helped me understand how to start a basic path down collecting data, and putting the
qualitative and quantitative data into meaningful explanations through charts and graphs as
well as understanding what those charts and graphs mean. Having the understanding of how
outliers can affect means, medians and standard deviations, as well as potentially skewing data.
There are so many statistics that are thrown around in the medical world, being able to work
through some of the data presented and understand if its important and factual or if its poorly
done.
Overall, Ive always been fascinated with statistics, and I think that taking this course and going
through this exercise has just touched the surface of my future studies around statistics. Its an
important tool in everyday life and decision making our brain probably computes more
statistics than we realize and it will be something Ill continue to improve as I continue my
schooling.

The Skittles Project Final

Uploaded by

Copyright:

Available Formats

The Skittles Project Final

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Skittles Project Final

Uploaded by

Copyright:

Available Formats

The Skittles Project

Skittles Sample Data Collection:

5 Number Summary for Skittles Sample Data Set:

You might also like